A social platform for motorsport, where lap telemetry becomes something you can share.
The problem
Lap telemetry arrives from hardware that agrees on almost nothing. A RaceBox unit, an MXTelemetry logger, a Columbus GPS, a SatellCom device and a phone app each produce a different sample rate, a different coordinate precision and a different idea of where a lap begins. Rendering four drivers on one corner means normalising all of it first, then drawing racing lines, braking points and acceleration zones over a map without the interface stuttering on a mid-range phone at trackside.
The approach
Ingest normalises every source into one canonical lap format at upload time, so nothing downstream needs to know which device produced a trace. Lap geometry is simplified server-side before it reaches the client — a full-rate GPS trace has far more points than a map needs at any given zoom. Sector detection is derived from the track geometry rather than trusting each device to mark its own start line, which is what makes laps from different hardware comparable at all.
The outcome
Drivers can put their lap next to someone else’s on the same corner regardless of what recorded it, which is the thing that turns a lap timer into a social product. Track data covers more than 170 international circuits, and the app ships in eight languages across iOS, macOS and Apple Vision.
What mattered
One lap format, many devices
RaceBox, MXTelemetry, Columbus and SatellCom traces normalise on ingest, so comparison logic never branches per device.
Geometry-derived sectors
Sector splits come from the circuit, not from each device’s own start line — the only way laps recorded on different hardware compare honestly.
Built for trackside
Traces are simplified before they reach the phone. Connectivity at a circuit is poor and the device is usually hot.